Cardiovascular Risk Prediction Using Machine Learning and Multidimensional Health Data

Abstract

Cardiovascular disease remains one of the leading causes of death worldwide, highlighting the need for reliable approaches to identify individuals at risk at an early stage. This study explores the use of machine learning techniques to predict cardiovascular risk using multidimensional population health data. The analysis is based on the 2023 Behavioral Risk Factor Surveillance System (BRFSS) dataset, which includes demographic information, health indicators, and lifestyle-related factors associated with cardiovascular outcomes. Several supervised learning models were implemented and compared, including Logistic Regression, Random Forest, XGBoost, LightGBM, and Artificial Neural Networks. These models were trained and evaluated to determine their effectiveness in identifying individuals with elevated cardiovascular risk. The results show that ensemble-based methods achieved stronger predictive performance, with LightGBM providing the most accurate classification in this study. To enhance transparency, explainable artificial intelligence techniques were applied to examine the key variables influencing model predictions. The findings highlight the potential of machine learning approaches to support early cardiovascular risk detection and contribute to more data-driven preventive healthcare strategies.